You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by John Wilcock <jo...@tradoc.fr> on 2004/10/14 11:18:45 UTC

[RD] incorrect real name (was Re: feeding frenzy for ws.surbl.org!!!)

On Thu, 14 Oct 2004 08:51:56 +0200, John Wilcock wrote:
> In particular, I have an idea for the To field:
> 
> | To: "judson burrows" <jo...@tradoc.fr>
> 
> My name isn't "judson burrows", or Ophelia Rrnyihie, or "edmond olivio",
> or any of the numerous similar names spammers have addressed me as
> recently! It ought to be fairly easy to write a rule specific to me that
> fires if the real name contains neither John nor Wilcock, 

I've just come up with a first attempt at this, that seems to work:

| header   __TO_NOREALNAME        To =~ /^["\s]*\<?\S+\@\S+\>?\s*$/
| header   __TO_GOODREALNAME_JOHN To =~ /^\"?john\s+wilcock\"?\s+\<?john\@\S+\>?\s*$/i
| header   __TO_JOHN              To =~ /john\@/i
| meta     local_BADREALNAME_JOHN __TO_JOHN && !__TO_GOODREALNAME_JOHN && !__TO_NOREALNAME
| describe local_BADREALNAME_JOHN To John but with wrong real name
| score    local_BADREALNAME_JOHN 2.0

but only for me personally, of course. Anyone any comments on whether
this rule could be improved or optimised? I'm sure it ought to be
possible to write a single regex rather than these three meta-rules...

And how about the more general case I mentioned:

> a generic plug-in that does an LDAP lookup or similar in order to
> determine the real name of J. Random User then checks whether the first
> and last names match? There's a challenge for you...

Any takers?

John.

-- 
-- Over 2500 webcams from ski resorts around the world - www.snoweye.com
-- Translate your technical documents and web pages    - www.tradoc.fr